When we build a kanban board to manage our work (either practicing Kanban or Scrum or something else) we usually create a Backlog list (usually the first column) and a To Do list (usually immediately following the Backlog). I’ve noticed that many times the separation between the two is artificial and people don’t always understand the critical difference between the two. I’d like to discuss it here.
The backlog is a list of things we think we should do, for many reasons (clients, our own needs etc.). Moreover, it is prioritized, so things on the top are things we really really want (or need) to do while things at the bottom are probably fantasies that will never happen.
For instance, the backlog of an airport is all the flights and the backlog of a road system is all the cars that need to go somewhere. And the backlog of a development team is all the software that needs to be developed.
As stuff goes up in priority it becomes more concrete and more focused, shedding off redundant things, like a baby growing up and losing their baby fat. This process is called backlog grooming or backlog refinement. And still the backlog is a list of things we want to do, we should do, we need to do – even the topmost items. Not more than that.
The To Do is a list of things we decided to do now because we believe that we can accomplish them in high quality and timely manner. This is a big decision. Moving something from the Backlog to To Do takes a lot of responsibility. You should consider it gravely. The hand holding the note, moving it from Backlog to To Do (or more likely, firmly grasping the mouse) should tremble.
Once you decide to do something, you want to do it very quickly. This is our cycle time. The shorter our cycle time is, the more focused we are, the chances for things going astray is lower, the chances for us changing our mind midway is smaller and many other good things – see Don Reinersten’s book Product Development Flow for the full list. Mr. Reinersten gives airplanes as an example. I’ll extend it a bit: airplanes should start boarding only when it is known that the airplane can freely taxi to the runway and then land directly without circling. Imagine how much time would be saved. And same goes for traffic over a road. Imagine a system that could tell you when to go out of the house so traffic won’t be congested.
When the team has available capacity (or in Scrum during the planning meeting) the team makes this decision. The team decides they will do something. If they think it is not yet ready they should keep it in the backlog. Ready for what? Ready for the way the team-machine knows to process stuff. The team calibrates the way they work and the stuff going into the machine should fit this.
Same goes for an organization now planning to do something in the next quarter – once they decide they want to do it they should do everything they can to make it progress very fast.
Of course there’s a trade off between how much you invest in an item when it’s in the backlog and later. Investing while it is in the backlog will increase predictability on one hand, but on the other hand, it will increase the amount of things you do in parallel which results in less focus and more waste due to context switching. You need to find the right balance.
And what if something is really big? We can’t do it really fast, but we do it in small pieces, small pieces that make sense. We do it so that once we’re done with a piece we have something complete (even a draft) that can provide us with information for how to continue.
I found that, as I have a degree in procrastination, doing things that take very little time is a miracle. I lose focus very fast so if I want to complete something I need to complete it fast. But so does an organization, right? How focused is your organization?
Once you practice doing things very fast, in the required quality, you actually complete stuff.
The problem starts when we mix the Backlog with the To Do, when things we should do somehow become things we do. People in a car want to get to their destination in the morning and immediately they start driving. Then we get congestion. Projects arrive to the company and we immediately start development. Then we get congestion.
But why does it happen? Why do we mix the Backlog and the To Do?
First, it is easier to just start. It is easier to manage. There is something you need to do and you just do it. It is easier to manage when you react. It is easier for every driver to go out of their parking and start driving rather than waiting for some sign to start driving.
Second, it feels much better to start doing things you should do. We are industrious people and when there’s a job to do we make a plan and start doing it. I remember as a kid I was told the best way to do homework was to do it the day you got it. It stands against our intuition not to start something we need to do.
We need to learn to make the differentiation between what we need to do, what we should do, what we want to do and what we are doing, between the Backlog and the To Do. This separation will help us to focus on what we do, complete it when we need it, in the right quality and eventually do more and better than we do today.